RII Track-4:@NASA: Enhancing Flood Detection and Mapping by Using PolSAR, Metaheuristic, and Deep Learning Algorithms

Floods are among the deadliest disasters and the most common and frequent natural hazards to life, property, the economy, and the environment worldwide. There is a vital need for an accurate and quick mapping of flooded areas to improve response to and recovery from floods, water resources management, effective land management, and risk mitigation efforts.

Given the more frequent extreme weather events, detection of flooded areas should be quick to expedite the disaster response and maximize the success of mitigation efforts. Currently, there is a knowledge gap in the accurate and fast mapping of flooded areas with high-resolution synthetic aperture radar (SAR) data. This Research Infrastructure Improvement Track-4 EPSCoR Research Fellows (RII Track-4) project will provide a fellowship to an Associate Professor and training for a postdoctoral fellow at the University of Hawaii.

This work will be conducted in collaboration with researchers at NASA Marshall Space Flight Center (MSFC). This fellowship provides a state-of-the-art framework for the precise and rapid detection of flooded areas with high-resolution SAR data. The results from this fellowship will be beneficial for a diverse group of stakeholders, including hydrologists, government agencies and contractors, insurance providers, land developers, community planners, and individuals living in flood-prone areas.

The educational activities are designed to motivate underrepresented high school and undergraduate students to pursue advanced degrees in Science, Technology, Engineering, and Mathematics (STEM) fields and improve graduate students’ education. Videos will be created and disseminated to broaden public awareness and appreciation of the benefits of this fellowship for underserved U.S. Pacific Islanders.

The main goal of this fellowship is to transform the Principal Investigator (PI)’s research trajectory and benefit his home institution and jurisdiction by pursuing four specific objectives: 1) Building a strong partnership with his host at the NASA Marshall Space Flight Center: The PI will make four collaborative visits to the host site to advance his project and create a long-term collaborative relationship with the host. The host will also visit the PI’s institution twice to co-advise the postdoctoral researcher and present a seminar for stakeholders; 2) Advancing the knowledge of SAR-based flood mapping by developing and validating state-of-the-art computational tools: The PI and postdoctoral researcher will process SAR images related to four flood events, find their optimum polarimetric SAR (PolSAR) decompositions via metaheuristic algorithms, and use the deep Residual U-Net model to detect flooded areas; 3) Promoting education of underserved students in the field of hydrology, remote sensing, and natural hazards: The PI will also engage underserved students in research activities related to this fellowship to enhance their critical thinking skills.

The expected outcomes are to prepare the next generation of scientists to use scientific computing tools in hydrology and natural hazards, and close the workforce gap for minority students in geoscience, and 4) Creating effective disaster response strategies during a flood: Coordinating a working group of stakeholders to create protocols by which the flood maps from this fellowship can be used by local and state emergency managers to mitigate the impacts of floods on underrepresented populations living in coastal communities.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.